Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Convolution Properties II01:17

Convolution Properties II

179
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
179
Convolution: Math, Graphics, and Discrete Signals01:24

Convolution: Math, Graphics, and Discrete Signals

240
In any LTI (Linear Time-Invariant) system, the convolution of two signals is denoted using a convolution operator, assuming all initial conditions are zero. The convolution integral can be divided into two parts: the zero-input or natural response and the zero-state or forced response, with t0 indicating the initial time.
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
240
Convolution Properties I01:20

Convolution Properties I

144
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
144
Deconvolution01:20

Deconvolution

146
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
146

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Modelling the risk of West Nile virus infection in seven European countries from published serological and case notification data, 2008 to 2022.

Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin·2026
Same author

CTransFuse: A Hybrid Transformer-CNN Framework for Precise Meniscus Segmentation and Lesion Identification.

IEEE journal of biomedical and health informatics·2026
Same author

GraphLooper: predicting chromatin loops based on hierarchical multi-view graph pooling method.

Briefings in bioinformatics·2026
Same author

Leveraging UAV hyperspectral imaging for crop physiology and biochemistry: A comprehensive review of feature extraction and selection methods.

Plant phenomics (Washington, D.C.)·2026
Same author

Graph-based understanding of isomeric diversity in complex dissolved organic matter.

Communications chemistry·2026
Same author

Artificial intelligence in microscopic hair imaging for scalp disorders: From image acquisition to clinical decisions.

Medical image analysis·2026
Same journal

DNA: Improving text-based person search through distillation learning, negated relation-aware learning, and augmented representation learning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

MCFusion-DDI: Multimodal cross-attention fusion of local-global features and latent drug associations for explainable DDI prediction.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Multiscale EEG feature fusion for recognizing 3D object shapes through active touch.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Fixed/preassigned-time stabilization and time-energy tradeoff analysis of delayed memristive reaction diffusion neural networks.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

The landscape of pruning for large language models: A systematic review and unified taxonomy.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Fourier multi-component and multi-layer neural networks: Unlocking high-frequency potential.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles
  1. Home
  3. Hdconv: Heterogeneous Kernel-based Dilated Convolutions.
  1. Home
  3. Hdconv: Heterogeneous Kernel-based Dilated Convolutions.

Related Experiment Video

Quantifying Intermembrane Distances with Serial Image Dilations
07:45

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

6.4K

HDConv: Heterogeneous kernel-based dilated convolutions.

Haigen Hu1, Chenghan Yu1, Qianwei Zhou1

  • 1College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, 310014, PR China; Key Laboratory of Visual Media Intelligent Processing Technology of Zhejiang Province, Hangzhou 310023, PR China.

Neural Networks : the Official Journal of the International Neural Network Society
|August 1, 2024

View abstract on PubMed

Summary
This summary is machine-generated.

A new heterogeneous dilated convolution (HDConv) overcomes the gridding problem in computer vision by using independent dilation rates. This approach enhances feature extraction for tasks like image segmentation and object detection.

Keywords:
Dilated convolutionHeterogeneous structureImage segmentationReceptive field

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.0K

Related Experiment Videos

Quantifying Intermembrane Distances with Serial Image Dilations
07:45

Quantifying Intermembrane Distances with Serial Image Dilations

Published on: September 28, 2018

6.4K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.0K

Area of Science:

  • Computer Vision
  • Deep Learning
  • Image Analysis

Background:

  • Dilated convolution expands receptive fields but suffers from the gridding problem, which disrupts information integrity.
  • The gridding issue in dilated convolutions arises from their isomorphic structure, leading to disconnected feature relevance.

Purpose of the Study:

  • To introduce a novel heterogeneous dilated convolution (HDConv) to mitigate the gridding problem.
  • To enhance feature extraction by incorporating multi-scale kernels and larger receptive fields.

Main Methods:

  • Proposed HDConv with independent dilation rates across grouped channels.
  • Explored various dilation rate combinations to optimize large receptive fields.
  • Integrated HDConv as a plug-and-play module into existing neural network architectures.

Main Results:

  • HDConv effectively resolves the gridding problem inherent in standard dilated convolutions.
  • Demonstrated competitive performance in image segmentation and object detection tasks on datasets like ADE20K, Cityscapes, and COCO.
  • Achieved strong results on the UESTC-COVID-19 medical imaging dataset.

Conclusions:

  • HDConv offers a viable solution to the gridding problem, improving information integrity in feature maps.
  • The proposed module shows significant potential for advancing computer vision applications, particularly in image segmentation.
  • HDConv is a versatile, plug-and-play component for enhancing existing deep learning models.